Text version
Delirium remains a huge driver of morbidity and mortality among hospitalized patients, leading to longer lengths of stay, increased time in intensive care unit settings, and worse outcomes. DSM-5-TR now includes a diagnosis of persistent delirium, recognizing that symptoms may persist for weeks to months in about 20%–30% of patients. Newer evidence also suggests that up to 5%–10% of patients never fully recover from delirium, essentially assuming a new worsened cognitive baseline.
Add to this the high percentage of patients who do recover cognitively but experience PTSD or post-ICU syndrome (PICS) as a result of their delirious experiences and you’ve got a major driver of morbidity and mortality. Unfortunately, when it comes to using medication to treat delirium, it’s been a series of disappointments. Several recent studies highlight something we’ve known for a while, that antipsychotic medication does not treat delirium or reduce the length of time that patients are delirious. Although they do still have a very important role in managing symptoms of delirium, including reducing harm to patients and staff and mitigating distressing perceptual disturbances like hallucinations, the mantra remains that the only real treatment for delirium is to treat the underlying cause. If only we could prevent delirium from happening in the first place.
I’m Scott Beach and this is Quick Takes for the Psychopharmacology Institute. Let’s look at a recent study in the Journal of Psychosomatic Research suggesting that the use of NSAIDs may be associated with a reduced risk of delirium. This retrospective cohort study examined 1,274 patients. It looked at several anti-inflammatory agents, including NSAIDs, aspirin, glucosamine, and others. About one-third of the patients in the studies used NSAIDs specifically. Prevalence of delirium in the NSAID group was 23% vs 35% in the non-NSAID group.
When the authors controlled for age, sex, dementia status, and hospital department—all important risk factors and confounders—the risk of delirium tended to be reduced by NSAID use, with an odds ratio of 0.76. When patients with and without dementia were examined separately using logistic regression, NSAID usage was associated with reduced delirium only for patients without dementia. NSAID use was also associated with a lower 1-year mortality risk. Findings were not statistically significant for other anti-inflammatory agents.
This idea that anti-inflammatory agents might be useful in delirium makes good intuitive sense. Whereas many of us were taught a very simplified model of delirium represented as an imbalance between dopamine and acetylcholine, more recent theories of delirium are markedly complex involving a host of factors, including neuronal aging, oxidative stress, neuroendocrine dysregulation, and circadian dysregulation. Neuroinflammation is considered one of the most important factors and is thought to play a major role in delirium. Think about it: We know that states of massive systemic inflammation carry high rates of delirium. These include things like long bone fractures, burn injuries, major surgeries like AAA repair, and more recently COVID-19 infection.
Higher inflammatory markers, including IL-6 and CRP, are typically found in patients with delirium. One theory is that these and other peripheral inflammatory cytokines induce microglia in the brain to release inflammatory cytokines centrally, leading to delirium. Alternatively, the peripheral inflammatory cytokines may also cross a leaky blood–brain barrier, which can be weakened by the inflammatory processes. Now, if there was a way to mitigate or block that peripheral inflammation, perhaps rates of delirium would decrease.
Prior studies have suggested that NSAIDs given with the specific intention of preventing postoperative delirium have shown some promise. This study goes a step further by suggesting that routine usage of NSAIDs for indications unrelated to delirium may also be helpful. Even recognizing some limitations to this study, which we’ll get to in a moment, these results are important.
When it comes to using medications to prevent delirium, very few things have been shown to work. If NSAIDs show some consistent promise in this regard, that could be a game changer. They’re relatively inexpensive, widely available, and generally well-tolerated. Importantly, though, they aren’t without side effects, especially for patients who may have risks for GI bleeding, clotting disorders, or kidney injury. NSAIDs are also being talked about as a way to mitigate risk and slow progression of dementia, particularly Alzheimer’s via their anti-inflammatory properties, though results are quite mixed at this stage.
As with all studies, it’s important to recognize the limitations here. The study was retrospective with diagnosis of dementia based on chart review. Rating scales were used to determine the presence of delirium with adjudication from a consultation-liaison psychiatrist, which is pretty good. The most obvious limitation, though, is that prescription history was used for all anti-inflammatory agents. So, if patients were taking over-the-counter aspirin or NSAIDs that were not captured in the EHR, these would not have been accounted for. Dose and duration of anti-inflammatory medication usage was also not controlled. So, we really don’t have a great sense of how much and how regularly people were taking these agents. And we can’t say for sure that everyone who had a prescription was taking it and everyone who didn’t have a prescription was not taking it.
What about the finding that NSAIDs only reduced delirium in patients without dementia but not in those with dementia? Well, this kind of makes sense. We know that dementia is itself a huge risk factor for delirium. We often talk about this as patients having a vulnerable substrate. When the brain is already running on fumes, it doesn’t take much of an insult to tip the scales into acute brain failure, which is essentially what delirium is. When I was in med school, we had a mnemonic for causes of delirium called I WATCH DEATH. I don’t think they teach that anymore because pretty much anything can cause delirium if the substrate is vulnerable enough. Even cases of delirium due to constipation or urinary retention, the so-called cystocerebral syndrome, have been reported. Perhaps dementia is such an important factor that NSAIDs really can’t overcome that vulnerability enough to be helpful.
Overall, though, these findings are still encouraging. We need larger prospective studies before prescribing NSAIDs to reduce delirium becomes common practice, especially given some of the risks, but this is a good start. Beyond their anti-inflammatory properties, NSAIDs can also reduce the use of opioids, which might be another advantage in terms of reducing delirium. Those of us who care for patients in general hospital settings should all stay tuned on this one to see how it plays out.
Abstract
Anti-Inflammatory Medication Use Associated With Reduced Delirium Risk and All-Cause Mortality: A Retrospective Cohort Study
Takehiko Yamanashi, Eleanor J Sullivan, Katie R Comp, Yoshitaka Nishizawa, Cade C Akers, Gloria Chang, Manisha Modukuri, Tammy Tran, Zoe-Ella E M Anderson, Pedro S Marra, Kaitlyn J Crutchley, Nadia E Wahba, Masaaki Iwata, Matthew D Karam, Nicolas O Noiseux, Hyunkeun R Cho, Gen Shinozaki
Objective: To investigate the relationship between history of anti-inflammatory medication use and delirium risk, as well as long-term mortality.
Methods: In this retrospective cohort study, subjects recruited between January 2016 and March 2020 were analyzed. Information about anti-inflammatory medication use history including aspirin, NSAIDs, glucosamine, and other anti-inflammatory drugs, was collected. Logistic regression analysis investigated the relationship between anti-inflammatory medications and delirium. Log-rank analysis and cox proportional hazards model investigated the relationship between anti-inflammatory medications and one-year mortality.
Results: The data from 1274 subjects were analyzed. The prevalence of delirium was significantly lower in subjects with NSAIDs usage (23.0%) than in those without NSAIDs usage (35.0%) (p < 0.001). Logistic regression analysis controlling for age, sex, dementia status, and hospitalization department showed that the risk of delirium tended to be reduced by a history of NSAIDs use (OR, 0.76 [95% CI, 0.55 to 1.03]). The one-year mortality in the subjects with NSAIDs (survival rate, 0.879 [95% CI, 0.845 to 0.906]) was significantly lower than in the subjects without NSAIDs (survival rate, 0.776 [95% CI, 0.746 to 0.803]) (p < 0.001). A history of NSAIDs use associated with the decreased risk of one-year mortality even after adjustment for age, sex, Charlson Comorbidity Index, delirium status, and hospitalization department (HR, 0.70 [95% CI, 0.51 to 0.96]).
Conclusion: This study suggested that NSAIDs usage was associated with decreased delirium prevalence and lower one-year mortality. The potential benefit of NSAIDs on delirium risk and mortality were shown.
Download PDF and other files
Reference
Yamanashi, T., Sullivan, E. J., Comp, K. R., Nishizawa, Y., Akers, C. C., Chang, G., … Shinozaki, G. (2023). Anti-inflammatory medication use associated with reduced delirium risk and all-cause mortality: A retrospective cohort study. Journal of Psychosomatic Research, 168, 111212.
